"Nicolai Tufar" <[EMAIL PROTECTED]> writes: >> We might think that the Turkish-locale problem Devrim Gunduz pointed out >> is a must-fix, too. But I'm not convinced yet what to do about it. > Here is a first try to fix what Devrim Gunduz talked about.
I still don't much like having a locale-specific wart in the parser (and the code you give could not work anyway --- for starters, the first argument of setlocale is not a pointer). A possible compromise is to apply ASCII downcasing (same as in keywords.c) for 7-bit-ASCII characters, and apply tolower() only for character codes above 127. In other words unsigned char ch = (unsigned char) ident[i]; if (ch >= 'A' && ch <= 'Z') ch += 'a' - 'A'; else if (ch > 127 && isupper(ch)) ch = tolower(ch); ident[i] = (char) ch; In reasonably sane locales this will have the same effects as currently, while in unsane locales it will ensure that basic-ASCII identifiers are treated the way we want. Comments? regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match